large language models

rimi. — Open Convention for Conversational Reliability of LLMs

An open, collaborative and testable convention that sets what a large language model does when it would otherwise guess, smooth over or invent: ambiguous answers, conflicting sources, missing data, and resource use. Conformance levels A, AA, AAA with a separate frugality indicator S1, S2, S3.

VisKnacks: Agentic ParaView Visualization on HPC

Artificial intelligence can assist in prototyping scientific visualizations. To assist visualization experts, we introduce VisKnacks, an autonomous ParaView visualization agent enabled by large language models (LLMs). VisKnacks integrates multi-agent orchestration, retrieval-augmented generation, and containerized rend...

Appendix to the article: "An Industrial Evaluation of DeepQuali: Assessing User Story Quality with Large Language Models"

Appendix to the article: “An Industrial Evaluation of DeepQuali: Assessing User Story Quality with Large Language Models”

LLM-CodeSmell-Detection-AppInventor

This repository contains the implementation, datasets, prompts, and experimental results used in the study on code smell detection in MIT App Inventor applications using local Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG).

Reproducibility Package for Comparative Evaluation of Lexical Machine Learning, CodeBERT Fine-Tuning, Prompt-Based LLMs, and Retrieval-Augmented Generation for SATD Detection

This archive contains the reproducibility package for the manuscript titled “Comparative Evaluation of Lexical Machine Learning, CodeBERT Fine-Tuning, Prompt-Based Large Language Models, and Retrieval-Augmented Generation for Self-Admitted Technical Debt Detection.” The package includes the executed notebook, source co...

The Standard Model Carrier Exhaustion Closure

Primitive Closure, Skin Openness, BSM Audit, and No Same-Scope Primitive Escape This paper completes the Standard Model carrier-exhaustion sequence. It asks whether any same-scope Standard Model-associated attribute retains primitive standing outside the certified carrier package after the gauge, matter, charge, coupli...

Dataset for: Beyond the Tailpipe: A Critical Review of Competing Visions of Net Zero Transport

This dataset contains the search record, screening decisions, model prompts and data extraction matrix for the critical review titled "Beyond the Tailpipe: A Critical Review of Competing Visions of Net Zero Transport." The review is a critical review conducted as a critical interpretive synthesis. It examines how the l...

Writing is thinking: toward a neuropsychological model of scientific writing as knowledge construction

Scientific writing has historically been conceptualized as a communicative act—the registration of knowledge elaborated through independent cognitive processes. This article challenges that assumption and proposes that scientific writing constitutes an active neurocognitive system of knowledge construction and a privil...

Leakage-Aware LLM Augmentation for Attrition Prediction: A DecisionCentric Evaluation

A leakage-aware, dual-network augmentation framework that integrates the semantic reasoning of Large Language Models with the distributional rigorousness of GAN discriminators is proposed, offering HR practitioners a robust, decision-centric tool to identify at-risk employees without sacrificing transparency.

Interpretable speech biomarkers of psychological resilience in major depressive disorder via expert-guided large language models

This framework successfully decodes cognitive coping strategies independent of depressive mood, providing potentially actionable targets for personalized psychiatric interventions and a promising, interpretable digital biomarker for psychological resilience.

Exploring young adult cancer survivors’ perspectives on generative AI chatbots for symptom support

Abstract Background Young adult cancer survivors (ages 18–39) often experience persistent treatment-related symptoms that negatively affect daily functioning, psychosocial well-being, and quality of life, yet report unmet educational needs for symptom management. Generative artificial intelligence (AI) technologies, su...

Benchmarking Stylometric Metrics for AI Authorship Verification: A Glass-Box Forensic Framework Across Language Models